Jan Ringoš

5.1K posts

Jan Ringoš banner
Jan Ringoš

Jan Ringoš

@JanRingos

TRIM CORE SOFTWARE s.r.o. @trimcore systems engineer and C++ programmer

Ostrava, Czech Republic Katılım Mayıs 2012
231 Takip Edilen184 Takipçiler
Sabitlenmiş Tweet
Jan Ringoš
Jan Ringoš@JanRingos·
Jan Ringoš tweet media
ZXX
1
0
3
0
Jan Ringoš
Jan Ringoš@JanRingos·
@dhh I've seen many extremely valuable and well-researched Wikipedia pages deleted for nothing but political reasons - usually thinly-veiled as not-notability or outright false questionability of sources. It's sad.
English
0
0
0
155
Jan Ringoš
Jan Ringoš@JanRingos·
All, SSE4.2, AVX and AVX2, ISA extensions are available in 32-bit mode too, but MSVC recognizes the new switch only for 64-bit target. You can still build 32-bit code with AVX. If you throw /arch:AVX, then it will generate SSE4.2 instruction in 32-bit code too. But it's not possible to set the cutoff at SSE4.2 which would be the most reasonable for 32-bit code today.
English
0
0
0
23
Jan Ringoš
Jan Ringoš@JanRingos·
Why is /arch:SSE4.2 only available for X64?
English
1
0
1
624
Jan Ringoš
Jan Ringoš@JanRingos·
@chainloaded @angelroom0 I've personally not bought several titles because of prominent cheating, both in multiplayer and challenge leaderboards, so I'd say it had real-life consequences to their income.
English
1
0
0
22
elias
elias@chainloaded·
@angelroom0 Drivers are supposed to enable compatibility with hardware and facilitate security. Preventing some kids from running cheats in a video game with 0 real life stakes at all does not justify running obfuscated kernel code that may endanger stability and tampers with the system
English
2
0
0
172
Jan Ringoš retweetledi
Wise
Wise@trikcode·
I haven't seen a C++ vibecoder yet. I wonder why?
English
774
234
7.2K
1.3M
Jan Ringoš
Jan Ringoš@JanRingos·
I'm currently working on a project that uses __vectorcall for a whole solution via /Gv, so I probably view it as more feasible and less problematic than it actually is. If MSVC would start inventing ideal calling conventions for every internal call, it would be awesome for performance, but probably absolute catastrophe for everything beginning with debuggers and ending with Detours. I'd still love it though. With APX, being as huge as it is, I almost expected Windows to come with a new kernel built with it. Treating it as just /feature:APX doesn't really feel like it's doing it justice.
English
0
0
0
12
Pedro Justo
Pedro Justo@itanium_guy·
@JanRingos @LinkedIn x86-64 does have alternate ABIs like verctorcall, but they are boutique ABIs, used in intimate contracts, not hyper-reusable API interfaces. Technically, compilers can make their own ABI (add registers) between transitive-closed calls to non-exposed APIs.
English
2
0
1
28
Jan Ringoš
Jan Ringoš@JanRingos·
@docsmsft Absolutely not. Registry is superior in all regards. The only thing missing is simple binary import/export of subkeys with proper support for ACL inheritance.
English
0
0
0
126
Microsoft Dev Docs
Microsoft Dev Docs@docsmsft·
If Windows were designed today, would the Registry still exist? Mark Russinovich breaks down how the Registry works—and why a more flexible, file-based configuration model might win out in a modern design. Registry info: msft.it/6012vpoK0
English
18
26
195
56.2K
Jan Ringoš
Jan Ringoš@JanRingos·
@m_schuetz I'd like to be able to write something like: auto x = 0u:16; auto y = 0u:4×32;
English
0
0
0
111
Markus Schütz
Markus Schütz@m_schuetz·
Bringing sanity back to C++
Markus Schütz tweet media
English
63
44
1.9K
113.8K
Jan Ringoš
Jan Ringoš@JanRingos·
Hmm, SEC_LARGE_PAGES don't count towards process commit, but MEM_LARGE_PAGES do.
English
0
0
0
42
Jan Ringoš
Jan Ringoš@JanRingos·
@JakubSzanto „Nevěřím žádné statistice, kterou si sám nezfalšuji.“
Čeština
0
0
0
29
Jan Ringoš
Jan Ringoš@JanRingos·
More often than not I've seen people steal clean memes and slap their watermark on it. I know a handful of mine got that treatment. So if they didn't know it was from you, perhaps it wasn't ill-intentioned action. But of course, reverse searching for the original is always the best action.
English
0
0
0
65
Jan Ringoš
Jan Ringoš@JanRingos·
@AnnatarTheMaia Real world data coming into a system aren't uniformly random, thus the low-quality output of std::rand is better match. Or do you mean C++ as a whole thing, lol?
English
1
0
0
13
Annatar
Annatar@AnnatarTheMaia·
@JanRingos And I firmly maintain, after having put the bread on the table programming in C++, that it is absolute garbage, no matter which additional, complicated gimmick is added to it. You can't polish a turd.
English
1
0
0
27
Jan Ringoš
Jan Ringoš@JanRingos·
And I firmly maintain that 'std::rand' is better than 'std::mt19937/std::uniform_int_distribution' for real-world comparable benchmarking.
English
1
0
2
532
Liberta Cherguia 🇪🇺
Liberta Cherguia 🇪🇺@MbarkCherguia·
Man minding his own business gets randomly menaced and harassed by this guy… What would you do in this situation?
English
1.4K
587
2.4K
107.3K
Windows Insider Program
Windows Insider Program@windowsinsider·
And we're live with all todays builds! Thank you Insiders for patience as we had a slight publishing delay, we hope you enjoy today's Taskbar and Start announcements! aka.ms/WIPTaskbarStart
English
17
42
261
23.8K
Jan Ringoš
Jan Ringoš@JanRingos·
@windowsinsider Can we get SDK for Experimental channel? Just as a peek under the hood.
English
1
0
0
296
Jan Ringoš
Jan Ringoš@JanRingos·
@Studio384 And allowed plain Win32 apps to use them, not lock them behind UWP managed packaging or whatever is it called.
English
0
0
1
47
Yannick
Yannick@Studio384·
Maybe a hot take, but Microsoft should never have killed of Live Tiles and just have expanded their capabilities and used them as widgets instead. Even if they made widgets pinable to the desktop, right now, they don't really have any widgets I'd want to use.
Xeno@XenoPanther

@zacbowden Wasn't there talk of widgets on the desktop a a year or two ago? Whatever happened to that

English
4
1
13
758